Modeling Provenance for Semantic Desktop Applications

As the volumes of digital resources grow exponentially, users face the threat of information overload. Almost everything we see, read, hear, write and measure is collected and made available via computational information systems (Carvalho et al. 2006). The problem is not so much finding information, but rather, developing computational solutions that help manage digital data in a meaningful way. In this paper we tackle this problem from the user’s perspective. We explore Semantic Desktop applications, which combine ontologies, taxonomies, and metadata in general to enhance information management and help reduce the difficulty of locating data stored in personal computers. We argue that such applications would benefit if endowed with the ability to autonomously harvest provenance metadata and index content accordingly and propose a generic provenance model for this purpose.